There has been conventionally known a headrest disclosed in Patent Literature 1 (Japanese Patent Laid-Open No. 2015-097682). The conventional headrest has acoustic speakers 125a and 125b that generate sound based on an audio signal, and an acoustic microphone 152 that collects sound and converts the sound into an electric signal, as denoted by reference sign 104 in FIG. 9, for example.
In the conventional headrest 104, a sound collection section of the acoustic microphone 152 and vibrating sections of the acoustic speakers 125a and 125b are disposed side by side on the surface of a core member 114. Consequently, the conventional headrest 104 has the problem that sound generated from the acoustic speakers 125a and 125b is collected by the acoustic microphone 152, and accuracy of voice recognition is reduced.